Filter paper method of soil suction measurement

Mtodo del papel de filtro para la medida de la succin del suelo

K. V. Bicalho
Federal University of Espirito Santo, Vitoria, ES, Brazil

A. Gomes Correia & S. Ferreira,

University of Minho, Guimares, Portugal

J-M Fleureau
Ecole Centrale de Paris, Chatenay-Malabry, France

F. A. M. Marinho
University of So Paulo, So Paulo, SP, Brazil

ABSTRACT
The capillary pressure of the soil (i.e., the pressure difference between air and water components in soil voids)
is a key variable in the analysis of the hydro-mechanical behavior of unsaturated soils. Therefore a simple and
economical laboratory method for the measurement of the capillary pressure of the soil (also known as soil
matric suction, the reference being the atmospheric pressure), even if a degree of approximation is involved,
is of considerable value. The filter paper method calculates soil suction indirectly by measuring the gravimetric water content of the filter paper at equilibrium that is related to soil suction through a predetermined calibration curve. The advantages of the method are simplicity, economy and reasonable accuracy. It can be used
to measure suctions from 10 to 30000 kPa. In this paper, the authors use the contact filter paper method for
matric suction measurements of an unsaturated compacted silty sand (formed by the weathering of granite)
which has been used as a building material for a road in the north of Portugal. The matric suctions inferred
from filter paper measurements depend on the calibration between the water content of the filter paper and
suction. Therefore, three calibration curves proposed at the literature (Chandler et al. 1992; ASTM D 5298;
and Oliveira & Marinho 2006) for the Whatman 42 filter paper are used to interpret the measured filter paper
gravimetric water contents. The results of these tests are compared to other techniques (i.e., tensiometers, and
the osmotic technique) used to measure or control the negative pore water pressure in the compacted soil
specimens and the results obtained are reasonably accurate.
RESUMEN
La presin capilar del suelo (es decir, la diferencia de la presin entre el aire y los componentes del agua en
vacos del suelo) es una variable llave en el anlisis del comportamiento hidromecnico de suelos no saturados. Un mtodo por lo tanto simple y econmico del laboratorio para la medida de la presin capilar del suelo
(tambin conocido como la succin matrica del suelo, la referencia que es la presin atmosfrica), mesmo si
un grado de aproximacin est implicado, es de valor considerable. El mtodo del papel de filtro calcula la
succin indirectamente utilizando curvas de calibracin. Las ventajas del mtodo son simplicidad, economa y
exactitud razonable. El mtodo del papel de filtro se puede utilizar para medir succiones a partir del 10 al
30000 kPa. En este artculo, los autores utilizan el mtodo del papel de filtro para la medida de la succin matric de una arena limosa compactada no saturada (formada por la meteorizacin del granito) que se ha utilizado como material de construccin para un camino en el norte de Portugal. Las succiones matric deducidas de
medidas del papel de filtro dependen de una calibracin entre el humedade del papel de filtro y la succin. Por
lo tanto, tres curvas de calibracin propuestas en la literatura (Chandler et al. 1992; ASTM D 5298; y Oliveira
& Marinho 2006) para el papel de filtro de Whatman 42 se utilizan para interpretar ls humedades gravimtricas medidas del papel de filtro. Los resultados de los ensayos se comparan a otras tcnicas (es decir, tensimetros, y la tcnica osmtica) usadas para medir o controlar la presin negativa en ls muestras compactadas
del suelo y los resultados obtenidos sea razonablemente exacto.
Keywords: laboratory tests, unsaturated soils, pore pressures, water retention curves

INTRODUCTION

A significant portion, if not the majority, of the

problems encountered in geo-environmental engineering practice involve unsaturated soils in the vadose zone (i.e., the unsaturated soil zone above the
water table). It is accepted that the matric suction,
which is commonly associated with the capillary
pressure (i.e., the pressure difference between air
and water components in soil voids, ua - uw), is a key
variable in the analysis of the hydro-mechanical behavior of unsaturated soils.
A comprehensive description of the experimental
techniques commonly used for measuring or controlling matric suction can be found in many references
(Fredlund & Rahardjo 1993; Lee & Wray 1995;
Ridley & Wray 1996; Lu & Likos 2004). The techniques vary widely in terms of cost, complexity, and
measurement range. The matric suction can be determined from previous calibration or can be measured directly. Because of the various difficulties involved in the direct measurements of the matric
suction, a simple and economical laboratory method
for the measurement of the capillary pressure of the
soil (also known as soil matric suction, the reference
being the atmospheric pressure), even if a degree of
approximation is involved, is of considerable value.
In this paper, the authors use the contact filter paper method for matric suction measurement of an
unsaturated compacted silty sand (formed by the
weathering of granite) which has been used as a
building material for a road in the north of Portugal.
The matric suctions inferred from filter paper measurements depend on a calibration between the water
content of the filter paper and suction. Therefore,
three calibration curves proposed at the literature
(Chandler et al. 1992; ASTM D 5298; and Oliveira
& Marinho 2006) for the Whatman 42 filter paper
are used to interpret the measured filter paper gravimetric water contents. The results of these tests are
compared to other techniques (i.e., tensiometers, and
the osmotic technique) used to measure or control
the negative pressure in the compacted soil specimens.
2

CONTACT FILTER PAPER TECHNIQUE

Filter paper technique was established for measuring soil suction by soil scientists and agronomists
(e.g., Gardner 1937; Fawcett & Collis-George 1967;
Al-Khafaf & Hanks 1974; and Hamblin 1981). In
geotechnical engineering fields, many researchers
have also used the technique as a routine method for
suction measurement (e.g., McKeen 1980; Chandler
& Gutierez 1986; Greacen et al. 1989; Chandler et
al. 1992; Ridley 1993; Marinho 1994; Houston et al.
1994; and Marinho & Oliveira 2006).
The filter paper method calculates the soil suction
indirectly from previous calibration. Basically, the

filter paper comes to equilibrium with the soil either

through vapor (total suction measurement) or liquid
(matric suction measurement) flow. At equilibrium,
the filter paper and the soil will have the same suction value. After equilibrium is established between
the filter paper and the soil, the gravimetric water
content of the filter paper disc is measured. The gravimetric water content of filter paper is converted to
suction using a calibration curve for the type of paper used. This is the basic approach suggested by the
American Society for Testing and Materials
(ASTM) standard D5298 for the measurement of either matric suction using the contact filter paper
technique or total suction using the non-contact filter
paper technique. The ASTM D 5298 employs a single calibration curve that has been used to infer both
total and matric suction measurements and recommends the filter papers to be initially oven-dried (16
h or overnight) and then allowed to cool to room
temperature in a desiccator. The ASTM D 5298
calibration curve is a combination of both wetting
and drying curves. However, because of the marked
hysteresis on wetting and drying of the filter paper,
the calibration curve for initially dry filter paper is
different from that of the initially wet filter paper.
Some publications presents calibration for the wetting path, with the paper initially air dry (Chandler
& Gutierez 1986; Chandler et al. 1992; Ridley 1993;
and Marinho 1994). Marinho & Oliveira (2006)
show that the calibration for the particular type of
paper is unique in relation to the type of suction (i.e.,
total or matric).
The contact filter paper technique is used for
measuring matric suction of soils. In the contact filter paper technique, water content of an initially dry
filter paper increases due to a flow of water in liquid
form from the soil to the filter paper until both come
into equilibrium. Therefore, a good contact between
the filter paper and the soil has to be established.
The contact filter paper method becomes inaccurate
in high matric suction range since water transport is
dominated by vapour transport (Fredlund et al.,
1995). The calibration curve for the filter paper matric suction measurement is commonly established
using a pressure plate apparatus (e.g., Al-Khafaf &
Hanks 1974; Hamblin 1981; Greacen et al. 1989).
It is important to note that only ash-less filter papers should be used in the filter paper technique. Although there are a number of ash-less filter papers
available, only Whatman 42 and Sleicher and
Schuell 59 (or SS 59) filter papers are commonly
used. Table 1 lists some calibrations presented in the
literature for the filter paper Whatman. 42. Most of
the curves are bilinear with an inflection point occurring at a filter paper gravimetric water content
value somewhere between 30 and 50 %. Figure 1
shows calibrations curves for Whatman 42 proposed
by Chandler et al. (1992), ASTM D 5298 and
Oliveira & Marinho (2006) for filter paper gravimet-

3
3.1

MATERIAL AND TEST PROCEDURES

Table 1. Calibrations curves for Whatman 42 filter paper

Reference
Suction W (%)
Log10 (suction)
range
(kPa)
ASTM D5298 Total
W <
5.327 -0.0779 w
and
45.3
Matric
ASTM D5298 Total
W >
2.412 -0.0135 w
and
45.3
Matric
Chandler &
Gutierrez
Matric
(*)
2.85 -0.0622 w
(1986)
Chandler et al. Matric W < 47
4.842-0.0622 w
(1992)
Chandler et al. Matric W > 47
6.050-2.48 Log w
(1992)
Oliveira &
Matric W < 33
4.83 0.0839w
Marinho
and
(2006)
Total
2.57 0.0154w
Oliveira &
Matric W > 33
and
Marinho
(2006)
Total
Note: w = Gravimetric water content
(*) suction range (80-6000 kPa)

2,5

Suction (log kPa)

ric water content (w) values ranging from 30 to 60

%. It should be noted the suction deviation between
Chandler et al. (1992), ASTM D 5298 and Oliveira
& Marinho (2006) increases for 30 % < w < 47 %
and the use of Chandler et al. (1992)s equation results in high suction values for these water content
values.
Likos & Lu (2002) conducted an analysis to
evaluate the accuracy and precision of total suction
measurement using the noncontact filter paper technique. They conclude that the filter paper calibration
curves can significantly vary among the same type
of filter paper from one batch or lot to another.
Therefore, they recommend batch-specific calibrations. It is important to mention that the non-contact
filter paper technique must be performed with extra
cares to avoid suction errors induced by temperature
gradient and relative humidity error. Figure 2 presents the results obtained by Likos & Lu (2002) for
seven different batches of Whatman 42 and the calibration curves proposed by Chandler et al. (1992)
and ASTM D 5298 for filter paper gravimetric water
content (w) values ranging from 0 to 40 %. The results indicate that the measurement deviation generally increases as w increases (i.e., as suction decreases) for w < 40 %. At relatively high values of
soil suction the use of Chandler et al (1992)s equation yields underestimated suction.
Marinho & Oliveira (2006) suggest that whenever the filter paper method is used as suction quantifier, one should check for the possibility of been
using a batch presenting a calibration curve that
differ from those frequently used in the scientific
community. It is suggested to make the calibration
of at least one point, verifying if that point is coherent with the calibrations proposed in literature.

1,5

ASTM D 5298

Chandler et al. (1992)

Oliveira & Marinho (2006)

0,5

30

35

40

45

50

55

60

Filter paper water content (%)

Figure 1. Calibrations curves for Whatman 42 filter paper for

w (%) values ranging from 30 to 60 %.

Test material

6
5,5
5
4,5

Suction (log kPa)

Tests were performed on a residual silty sand, hereafter called Perafita sand, resulting from weathered
granite, which has been used as a building material
for a road in the north of Portugal. Its grain size distribution is indicated in Figure 3. It contains about
20% of grains smaller than 80 m, with a layered
structure similar to that of clay particles. The liquid
limit of the Perafita sand is 32.6 %, the plastic limit
is 25 %, clay fraction is 2.5%, specific gravity is
2.66, standard Proctor optimum water content is
17.6% and the corresponding dry density is 16.8
kN/m3, modified Proctor optimum water content is
13.2% and the corresponding dry density is 18.6
kN/m3.

4
3,5
3

ASTM D 5298

2,5
2

Likos & Lu ( 2002)

(7 different batches)

1,5
1

Chandler et al. (1992)

0,5
0
0

10

15

20

25

30

35

40

Filter paper water content (%)

Figure 2. Calibrations curves for w (%) values ranging from 0

to 40 % for seven batches of Whatman 42 filter paper.

Passing (%)

80

d60/d10=17

60

2m

80m

40
20
0
0.001

0.01

0.1

10

Grains diameter (mm)

Figure 3. Grain size distribution of Perafita sand (Fleureau et

al. 2002)

3.2

Test program

The preparation procedure of samples is the same

for all the tests: the soil is sieved to avoid the presence of coarse grains (maximum size 4.75 mm), then
it is mixed up with the right quantity of water; after
that, it is placed in a sealed plastic bag for 24 hours
to allow the hydric equilibrium to establish. The
contact filter paper tests were carried out on soil
specimens compacted to the Modified Proctor Optimum water content (13.2%) and nearly maximum
density (18.6 kN/m3). The compacted soil specimen
sizes were 102 mm in diameter and 23.35 mm high.
The test procedure involves placing a piece of initially air dry filter paper against the compacted soil
specimen whose matric suction is required and sealing the whole to prevent evaporation. The filter paper then wets up to a water content in equilibrium
with the magnitude of the soil matric suction, and
careful measurement of the water content of the filter-paper enables the soil matric suction to be obtained from a previously established correlation.
This provides a measure of the matric suction, which
is assumed to be the same numerically as the capillary pressure (the reference being the atmospheric
pressure).The Whatman 42 filter paper was used in
all tests.
The other techniques (i.e., tensiometers, and the
osmotic technique) used to measure or control the
negative pore water pressure in the compacted soil
specimens are not discussed in this paper since the
purpose herein is to discuss the filter paper technique only. Details of the experimental techniques
are given in Fleureau et al. (2002).
4

(uw - uatm). The volumetric water content is defined

as the soil porosity times the degree of saturation.
In order to verify the effect of the filter paper
calibration curves on the contact filter paper method
for matric suction measurement, the authors have
used three calibration curves proposed at the literature (Chandler et al. 1992; ASTM D 5298; and
Oliveira & Marinho 2006) to interpret the measured
contact filter paper gravimetric water contents (43 %
< w < 57%). It can be seen from Figure 4 that the
matric suctions inferred from filter paper measurements depend on a calibration between the water
content of the filter paper and suction and the use of
ASTM D 5298s equation underestimated suction
values for w > 45.5 %. As observed in Figure 1, the
difference in matric suction values based on either
Chandler et al. (1992)s equation or Oliveira &
Marinho (2006)s equation is small for 47 % < w <
60 %.
Based on the data in Figure 4 the matric suction
values inferred from filter paper measurements depend on the calibration between the water content of
the particular type of paper used and suction. The results suggest that the matric suction values based on
Chandler et al. (1992) and Oliveira & Marinho
(2006) calibration equations are reasonably well located on the drying-wetting path defined by the
other techniques (i.e., tensiometers, and the osmotic
technique) used by Fleureau et al. (2002) to measure
or control the negative pore water pressure in the
compacted Perafita sand specimens.
10000

Matric suction (kPa)

100

Fleureau et al. (2002)

Filter paper method
(Chandler et al. 1992)
Filter paper method
(ASTM D 5298)

1000

Filter paper method

(Oliveira & Marinho 2006)

100

10
0

10

15

20

25

30

35

Volumetric water content (%)

Figure 4. The measured matric suction values versus volumetric water content for compacted Perafita sand specimens.

TEST RESULTS AND ANALYSIS

The measured matric suction values of compacted

Perafita sand specimens resulting from several methods used by Fleureau et al. (2002) to control or
measure the matric suction and contact filter paper
tests investigated in this paper are plotted versus
volumetric water content in Figure 4. The term matric suction is used to indicate the negative pressure
of water relative to atmospheric air pressure, i.e. -

5 CONCLUSIONS
Matric suction measurements of unsaturated compacted silty sand specimens using the contact filter
paper method have been presented. The matric suctions inferred from filter paper measurements depend on a calibration between the water content of
the filter paper and suction. Experimental errors in-

duced by using a calibration curve that differ from

those frequently used in the scientific community
that may influence the experimental results have
been discussed.
There is a general agreement between the measured matric suction values of compacted Perafita
sand specimens resulting from several methods used
by Fleureau et al. (2002) to control or measure the
matric suction (i.e., tensiometers, and the osmotic
technique) and contact filter paper tests investigated
in this paper.
REFERENCES
Al-Khafaf, S. & Hanks, R. J. 1974. Evaluation of the filter paper method for estimating soil water potential. Soil Sci.
117, 194-199.
ASTM 1992. Standard test method for measurement of soil potential (suction) using filter paper. (D. 5298-92). Annual
Book of ASTM Standards, vol. 15.09.
Chandler, R. J., Crilley, M. S. & Montgomery- Smith, G. 1992.
A low-cost method of assessing clay desiccation for lowrise buildings. Proc. Instn. Civ. Engrs Civ. Engng. 92, May,
82-89.
Chandler, R. J. & Gutierrez, C. I. 1986. The filter paper method
of suction measurement. Geotechnique 36, No. 2,265-268.
Fawcett, R. G. & Collis-George, N. 1967. A filter paper
method for determining the moisture characteristics of soil.
Amt. J. Exp. Agric. Anim. Husb. 7, 162-167.
Fleureau, J.-M., Hadiwardoyo, S., Dufour-Laridan, E. Langlois, V, and Gomes Correia, A., 2002, Influence of suction
on the dynamic properties of a silty sand. Unsaturated
Soils, Juc, de Campos and Marinho (eds), Sweets and
Zeitlinger, Lisse, 2, 463-471.
Fredlund, D. G. & Rahardjo H. 1993. Soil mechanics for unsaturated soils. Wiley & Sons, New York.
Gardner, R. 1937. A method of measuring the capillary tension
of soil moisture over a wide moisture range. Soil Sci. 43,
277-283.
Greacen, E. L, Walker, G. R. and Cook P. G., 1989, Procedure
for the Filter paper method of measuring soil water suction,
Division of soils, Report 108,CSIRO Division of Water Resources, Glen Osmond, Australia.
Hamblin, A. P. 1981. Filter paper method for routine measurement of field water potential. J. Hydrol. 53, 355-360.
Houston, S. L. Houston, W. N. Wagner A. 1994. Laboratory
filter paper suction measurements, Geotechnical Testing
Journal, 17, 2, 185-194.
Lee, H. c. & Wray, W. K. 1995. Techniques to evaluate soil
suction a vital unsaturated soil water variable. 1 International conference on unsaturated soils, Paris, 615-622.
Likos, W. J. & Lu, N. 2002. Filter paper technique for measuring total soil suction.
Transportation Research Record
1786, Paper N0.. 02-2140, 120-128.
Lu, N., and Likos, W.J. 2004. Unsaturated soil Mechanics,
John Wiley and Sons, 556 p.
Marinho, F. A. M. 1994. Shrinkage behavior of some plastic
clays. PhD Thesis, Imperial College, University of London.
Marinho, F. A. M. & Oliveira, O. M. 2006. The filter paper
method revised. ASTM geotechnical testing journal, USA,
29 (3), 250-258.
McKeen, R. G. 1980. Field studies of airport pavements on expansive clay. Proc. 4th Int. Co& Expansive Soils 1, 242-26
I.

McQueen, I. S. & Miller, R. F. 1968. Calibration and evaluation of a wide-range gravimetric method for measuring
moisture stress. Soil Sci. 106, 225-23 1.
Oliveira, O. M. & Marinho, F. A. M. 2006. Evaluation of filter
paper calibration. UNSAT 2006.
Ridley, A. M. Dineen, K. Burland, J. B. & Vaughan, P. R.
2003. Soil matrix suction: some examples of its measurement and application in geotechnical engineering. Gotechnique 53, N.2, 241-253.
Ridley, A. M. 1993. The measurement of soil moisture suction.
PhD thesis, University of London.
Ridley, A. M. and Wray, W. K. 1996.Suction measurementA
review of current theory and practices. Proc., 1st Int. Conf.
on Unsaturated Soils, 1293-1322.